The thermal capacity of buildings enables heating, ventilating, and air-conditioning (HVAC) systems to be exploited as demand response (DR) resources. Optimal DR of HVAC units is challenging, particularly for multi-zone buildings, because this requires detailed physics-based models of zonal temperature variations for HVAC system operation and building thermal conditions. Using supervised learning (SL), this paper proposes a new strategy for optimal DR of an HVAC system in a multi-zone building. Artificial neural networks (ANNs) are trained with data obtained under normal building operating conditions. The ANNs are replicated using piecewise linear equations, which are explicitly integrated into an optimization problem for price-based DR. The problem is solved for various electricity prices and building thermal conditions. The solutions are used to train a deep neural network (DNN) to directly determine the optimal DR schedule, which is termed as SL-aided meta-prediction (SLAMP) here; the DNN can work as a price-and-optimal-demand curve. Case studies are performed using three different methods: explicit ANN replication, SLAMP, and physics-based modeling. The case study results verify that the proposed SL-based strategy is effective in terms of both practical applicability and computational time, while also ensuring the thermal comfort of occupants and cost-effective operation of the HVAC system.

In this paper, a new strategy for service restoration of de-energized loads is proposed using advanced technologies of microgrids (MGs) that accommodate distributed energy resources (DERs). In particular, backup distributed generators (DGs), initially used for critical loads within a building, have recently been operated under normal grid conditions while exporting excess power to other loads on the same or different feeders. In addition, smart switches (SSWs) have been installed in a distribution grid to reduce the frequency and duration of power outages. Using flexible communication links, the sensors of the SSWs exchange control and measurement signals with grid management systems, namely, an advanced distribution management system, an MG energy management system, and a DER management system. This paper focuses on developing a methodological framework to determine the operating modes of synchronous-machine DGs and the on-off operation of SSWs, so as to change dynamically the boundaries of MGs that are formed in a distribution grid for fault isolation and load restoration. Simulation case studies demonstrate that the proposed strategy is effective in mitigating the influence of a network fault and restoring de-energized loads while successfully maintaining frequency and voltage levels in MGs.

Thermal energy capacity of buildings can be coupled to power networks via heating, ventilating, and air-conditioning (HVAC) units. Optimizing the operation of HVAC systems in multizone buildings is a challenging task, as occupants have different thermal preferences dependent on time-varying indoor and outdoor environments. To estimate the social cost of demand response (DR), building aggregators need to assess occupants' thermal discomfort (TD), which is related to their productivity outcomes. This paper proposes a price-based DR strategy for multizone office buildings to co-optimize the energy cost of HVAC units and the TD levels of occupants. To overcome simplified TD representations, a mobile interface is developed that grants occupants the ability to indicate their personal thermal preferences. These preferences are modeled using artificial neural networks, which are explicitly and directly integrated into the optimal DR scheduling. In addition, we evaluate the thermal response of a multizone office building to the operation of a variable speed heat pump (VSHP). Using the models of occupants' TD and building thermal response, the optimization problem for the proposed DR strategy is formulated and solved with mixed-integer linear programming. The case study results verify that the proposed strategy successfully optimizes VSHP operations and occupants' TD levels, mitigating the risk of occupant interruption to the optimal DR schedule.

A new control method for a line-commutated converter-based (LCC) high-voltage direct-current (HVDC) system is presented and compared to a conventional strategy. In the proposed method, both the DC voltage and current of an LCC HVDC system are regulated to increase the short-term operating margin of DC power transfer and improve transient responses to DC power references. In particular, an increased operating margin of DC power transfer is achieved via the DC voltage regulation method. To verify the effectiveness of the proposed method, a state space model of an LCC HVDC system is developed considering DC voltage and current references as input variables and analyzed for various values of the DC line inductance and converter controller gains. The state space model can be used for time-efficient analyses of the dynamic characteristics of an LCC HVDC system. Simulation case studies are performed using MATLAB, where the state space model of the Jeju-Haenam HVDC system is implemented as a test case and compared to its comprehensive PSCAD model. The case study results suggest that the proposed method increases the short-term operating margin and speeds up the transient response of the HVDC system. Therefore, it will effectively improve real-time grid frequency regulation.

Unbalanced phase voltages in electricity networks prevent efficient and stable delivery of power and energy services from utility companies to end-users, and vice versa. While variations in voltage magnitude and power loss have been studied extensively, little attention has been paid to the development of linearized expressions to account for the variation in voltage unbalance. Using a standard definition of the voltage unbalance factor (VUF), this study is the first, to the best knowledge of the authors, in which a VUF sensitivity matrix is mathematically developed and analyzed with respect to variations in three-phase (3-ph) voltages and, consequently, 3-ph unbalanced load demand. To demonstrate its utility, we performed simulation case studies and compared the estimates made using the matrix to those from a power flow calculation algorithm. The VUF sensitivity matrix is expected to have a variety of practical applications for power systems, such as optimal scheduling and real-time control, which enables improvement in network power quality.

Due to the increasing market share of electric vehicles (EVs), the optimal thermal management (TM) of batteries has recently received significant attention. Optimal battery temperature control is challenging, requiring a detailed model and numerous parameters of the TM system, which includes fans, pumps, compressors, and heat exchangers. This paper proposes a supervised learning strategy for the optimal operation of the TM system in an EV. Specifically, for TM subsystems, individual artificial neural networks (ANNs) are implemented and trained with data obtained under normal EV driving conditions. The ANNs are then interconnected based on the physical configuration of the TM system. The trained ANNs are replicated using piecewise linear equations, which can be explicitly integrated into an optimization problem for optimal TM scheduling. This approach enables the application of a mixed-integer linear programming solver to the problem, ensuring the optimality of the solution. Simulation case studies are performed for the two operating modes of the TM system: i.e., integrated and separate modes. The case study results demonstrate that the ANN-based model successfully reflects the operating characteristics of the TM system, enabling accurate battery temperature estimation. The proposed optimal TM strategy using the ANN-based model is verified as effective in reducing the total energy consumption, while maintaining the battery temperature within an acceptable range.